Combining characters
~~~~~~~~~~~~~~~~~~~~
Handling is completely broken at this moment.
Type 1 fonts lack relevant data completely, and fontsets can't read TTF tables directly yet.

Current assumption is that showing /a /acutecomb /b yields something resembling áb.
This requires combining characters to have zero width and negative coordinates.
Some fonts do implement this, some don't. FreeMono in particular had this changed somewhere
between 2006 and 2008, with current versions having negative coordinates but non-zero width.

To handle current FreeMono version, u2ps 0.9 adds a dirty hack, enforcing zero-width
for glyphs it considers to be combining. It's better than nothing, but do not expect
good typesetting.


tcedilla/tcommaaccent/tcommabelow
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
For reasons beyond my understanding, proper names for these glyphs in FreeMono are:
	scedilla ş	scommaaccent ș
but
	tcedilla ţ	tcommabelow ț
This is while tcommaaccent is defined as an alias for tcedilla, a different glyph!
To make things even more interesting, scommabelow is defined (as an alias for scommaaccent)
but Scommabelow is not.

Unicode standard lists these characters are "[SsTt] with comma below" "[SsTt] with cedilla".
In the original Adobe Glyph List it's uniformly /[SsTt]commaaccent, however with
0x0162, 0x163 for t (cedilla, ţ) and 0x0218, 0x0219 for s (comma below, ș).

I suspect this may be wrong and will probably cause issues with fonts other than FreeMono,
but at least it works with FreeMono which is the default font for u2ps anyway.
Nimbus Courier/Times substitutes from gs package lack these glyphs altogether.

In case this has to be fixed on-site, it's all in res/Unidata/Agl,
and there is no simple way for changing it on per-font basis short of editing the font atm.

Rerefence:
ftp://ftp.unicode.org/Public/5.0.0/ucd/UnicodeData.txt
http://partners.adobe.com/public/developer/en/opentype/glyphlist.txt
